Jiri Slaby [Fri, 18 Jun 2021 06:15:07 +0000 (08:15 +0200)]
mxser: decrypt FCR values
Currently, some of the values written to FCR are magic constants. But
they are composed of well-defined bits. Use these named macros in place
of the constants.
Jiri Slaby [Fri, 18 Jun 2021 06:15:05 +0000 (08:15 +0200)]
mxser: introduce and use start/stop_tx helpers
Starting and stopping TX is performed on many places in the driver. It
is repeated outb and info->IER modification. Create few helpers
(__mxser_start_tx, mxser_start_tx, __mxser_stop_tx) and use them in the
code instead of open coding.
This should be a base for an uart driver for this device in the future.
Jiri Slaby [Fri, 18 Jun 2021 06:15:03 +0000 (08:15 +0200)]
mxser: don't start TX from tty_operations::put_char
TX is not supposed to be started from tty_operations::put_char. Line
disciplines ought to call tty_operations::put_char several times and
then do a single tty_operations::flush_chars to start TX.
Fix this in mxser by removing TX startup from mxser_put_char.
Jiri Slaby [Fri, 18 Jun 2021 06:15:02 +0000 (08:15 +0200)]
mxser: cleanup mxser_dtr_rts
mxser_dtr_rts now does all inb, outb, & or | in a single statement. The
code is hard to follow, so make it more readable by doing one-line =
one-statement.
Jiri Slaby [Fri, 18 Jun 2021 06:15:01 +0000 (08:15 +0200)]
mxser: cleanup mxser_rs_break
mxser_rs_break now does all inb, outb, & or | in a single statement. The
code is hard to follow, so make it more readable by doing one-line =
one-statement.
Jiri Slaby [Fri, 18 Jun 2021 06:14:58 +0000 (08:14 +0200)]
mxser: alloc struct mxser_board dynamically
There is no need to preallocate an array of four struct mxser_board's.
In most cases a single board or two will be present in a machine. So
allocate struct mxser_board as needed in ->probe.
This makes mxser_boards a bit array. There we store which indexes are
free (unallocated).
Jiri Slaby [Fri, 18 Jun 2021 06:14:56 +0000 (08:14 +0200)]
mxser: rework and simplify mxser_open
In mxser_open, tty->port is already set thanks to
tty_port_register_device in ->probe. So we can simply use container_of
to deduce the struct mxser_port's pointer. Thus avoiding divisions and
complex computations.
Jiri Slaby [Fri, 18 Jun 2021 06:14:55 +0000 (08:14 +0200)]
mxser: cleanup mxser_process_txrx_fifo
Rename process_txrx_fifo to mxser_process_txrx_fifo and:
* remove useless parentheses
* return from the 'if's true branch and process the rest in normal code
flow (shift the code one level left)
Jiri Slaby [Fri, 18 Jun 2021 06:14:54 +0000 (08:14 +0200)]
mxser: remove mxser_cardinfo
Now that mxser_cards array (of struct mxser_cardinfo) contains only port
count and a quirk, we can encode these two directly into
pci_device_id::driver_data. And we can remove this struct completely.
Jiri Slaby [Fri, 18 Jun 2021 06:14:53 +0000 (08:14 +0200)]
mxser: remove info message from probe
There is a dev_info in ->probe which prints device's name, bus and slot
numbers. All these can be fetched from the lspci output. So remove this
useless print.
This eliminates the whole mxser_cardinfo::name field.
Jiri Slaby [Fri, 18 Jun 2021 06:14:51 +0000 (08:14 +0200)]
mxser: pci, switch to managed resources
Switch to managed resources for PCI using pcim_enable_device.
Regions and device disabling is now taken care of by the core and we
need not unwind. Neither in ->probe, nor in ->remove. That simplifies
the code.
Jiri Slaby [Fri, 18 Jun 2021 06:14:48 +0000 (08:14 +0200)]
mxser: clean up the rest of MUST helpers
Clean up the rest of MUST helpers. They all can use the newly added
mxser_must_set_EFR. And passing a bool instead of having two functions
(_enable+_disable) simplifies the code a lot too.
Jiri Slaby [Fri, 18 Jun 2021 06:14:47 +0000 (08:14 +0200)]
mxser: introduce mxser_must_select_bank and use it
Again, selecting a bank in the EFR register is duplicated many times.
Extract it to a separate function (mxser_must_select_bank) and use it on
all the places.
This cleans up most of the helpers. And mxser_get_must_hardware_id now
returns the HW ID directly, not via a by-reference parameter.
Jiri Slaby [Fri, 18 Jun 2021 06:14:46 +0000 (08:14 +0200)]
mxser: dedup mxser_must_set_enhance_mode
There are multiple copies of setting the EFR register. Separate it to a
new universal function called __mxser_must_set_EFR. And use it in the
code by introducing mxser_must_set_enhance_mode instead of
mxser_disable_must_enchance_mode and mxser_enable_must_enchance_mode.
Jiri Slaby [Fri, 18 Jun 2021 06:14:45 +0000 (08:14 +0200)]
mxser: simplify mxser_ioctl_op_mode
* ModeMask local array is just inverted OP_MODE_MASK shifted by shiftbit.
Drop this array and use the value directly.
* return from the if's true branch and drop 'else' branch by moving the
code completely outside the if.
Jiri Slaby [Fri, 18 Jun 2021 06:14:44 +0000 (08:14 +0200)]
mxser: extract mxser_ioctl_op_mode
This ioctl code sets RS232, RS422 or RS485 mode. Move the code to a
separate function (mxser_ioctl_op_mode) with a note that these
mxser-special ioctls should be likely replaced eventually by TIOCGRS485
and TIOCSRS485.
Jiri Slaby [Fri, 18 Jun 2021 06:14:38 +0000 (08:14 +0200)]
mxser: don't allocate MXSER_PORTS + 1
The MXSER_PORTS's tty_device is never registered and neither its
tty_port instance exists. Hence, it's quite pointless to allocate it.
It used to be used for global information fetches via ioctls. We have
just removed these as non-existent tty_device+tty_port was exactly the
reason to remove the ioctls.
It would be peculiar to add this "virtual" port as we require a backing
tty_port since commit ecbbfd44a08f (TTY: move tty buffers to tty_port)
in 2012.
Jiri Slaby [Fri, 18 Jun 2021 06:14:37 +0000 (08:14 +0200)]
mxser: remove cnt from mxser_receive_chars
After the previous ioctls removal, cnt is needed only in
mxser_receive_chars_old now. So remove it from mxser_receive_chars and
mxser_receive_chars_new and account only in mxser_receive_chars_old.
Jiri Slaby [Fri, 18 Jun 2021 06:14:36 +0000 (08:14 +0200)]
mxser: remove MOXA_HighSpeedOn ioctl
1) The semantics are defined exactly nowhere.
2) The ioctl definition was never exposed to userspace, so noone should
actually be reliably using this.
3) It's not even mentioned in any of the mxser utilities.
Provided the above, remove this mxser-special ioctl.
Jiri Slaby [Fri, 18 Jun 2021 06:14:35 +0000 (08:14 +0200)]
mxser: remove MOXA_ASPP_OQUEUE ioctl
1) The semantics are defined exactly nowhere.
2) The ioctl definition was never exposed to userspace, so noone should
actually be reliably using this.
3) It's not even mentioned in any of the mxser utilities.
Provided the above, remove this mxser-special ioctl.
Jiri Slaby [Fri, 18 Jun 2021 06:14:34 +0000 (08:14 +0200)]
mxser: remove MOXA_GETMSTATUS ioctl
1) MOXA_GETMSTATUS is broken at least since commit ecbbfd44a08f (TTY:
move tty buffers to tty_port) in 2012.
Even though it's mentioned in the mon_pa moxa utility, obviously
noone ever tried that in that timeframe.
2) The ioctl definition was never exposed to userspace, so noone
should actually be reliably using this.
Provided the above, remove this mxser-special ioctl.
And remove the whole concept of mxser_ioctl_special as this was the last
ioctl in there.
Jiri Slaby [Fri, 18 Jun 2021 06:14:33 +0000 (08:14 +0200)]
mxser: remove MOXA_GETDATACOUNT ioctl
1) MOXA_GETDATACOUNT is broken at least since commit ecbbfd44a08f (TTY:
move tty buffers to tty_port) in 2012.
Even though it's mentioned in the mon_p moxa utility, obviously noone
ever tried that in that timeframe.
2) The ioctl definition was never exposed to userspace, so noone should
actually be reliably using this.
Provided the above, remove this mxser-special ioctl.
Jiri Slaby [Fri, 18 Jun 2021 06:14:32 +0000 (08:14 +0200)]
mxser: remove MOXA_CHKPORTENABLE ioctl
1) MOXA_CHKPORTENABLE is broken at least since commit ecbbfd44a08f (TTY:
move tty buffers to tty_port) in 2012.
2) The ioctl definition was never exposed to userspace, so noone should
actually be reliably using this.
3) It's not even mentioned in any of the mxser utilities.
Provided the above, remove this mxser-special ioctl.
Jiri Slaby [Fri, 18 Jun 2021 06:14:31 +0000 (08:14 +0200)]
mxser: remove MOXA_ASPP_LSTATUS ioctl
1) The semantics are defined exactly nowhere.
2) The ioctl definition was never exposed to userspace, so noone should
actually be reliably using this.
3) It's not even mentioned in any of the mxser utilities.
Provided the above, remove this mxser-special ioctl.
Jiri Slaby [Fri, 18 Jun 2021 06:14:30 +0000 (08:14 +0200)]
mxser: remove MOXA_ASPP_MON and friends
1) MOXA_ASPP_MON_EXT is broken at least since commit ecbbfd44a08f (TTY:
move tty buffers to tty_port) in 2012.
2) The ioctl definitions were never exposed to userspace, so noone should
actually be reliably using them.
3) They're not even mentioned in any of the mxser utilities.
So remove all these mxser-special ioctls: MOXA_ASPP_MON,
MOXA_ASPP_MON_EXT, and MOXA_SDS_RSTICOUNTER.
Jiri Slaby [Fri, 18 Jun 2021 06:14:29 +0000 (08:14 +0200)]
mxser: remove MOXA_SET_BAUD_METHOD ioctl
1) The semantics are defined exactly nowhere.
2) The ioctl definition was never exposed to userspace, so noone should
actually be reliably using this.
3) It's not even mentioned in any of the mxser utilities.
Provided the above, remove this mxser-special ioctl.
Jiri Slaby [Fri, 18 Jun 2021 06:14:25 +0000 (08:14 +0200)]
mxser: make xmit ring buffer variables unsigned
Make the xmit ring buffer pointers and counter (xmit_head, xmit_tail,
xmit_cnt) unsigned. Now, the assumptions for the compiler are clear --
they cannot be negative.
Jiri Slaby [Fri, 18 Jun 2021 06:14:24 +0000 (08:14 +0200)]
mxser: correct types for uart variables
In struct mxser_port, all the UART registers and masks, and the xon/xoff
character are stored into int. Let's re-type all these as UART registers
are 8bit (u8) and xon/xoff character is an unsigned char (cc_t in
ktermios).
This save some bytes in memory, but more importantly the change makes it
what it really is.
Jiri Slaby [Fri, 18 Jun 2021 06:14:21 +0000 (08:14 +0200)]
mxser: extract mxser_receive_chars_new
Extract the new handling of receive chars (using the MUST chip
enhancements) into a separate function. This eliminates the need for the
intr_old label and makes the code of mxser_receive_chars more compact.
In the next step, we will extract the old handling too.
Jiri Slaby [Fri, 18 Jun 2021 06:14:20 +0000 (08:14 +0200)]
mxser: simplify mxser_interrupt and drop mxser_board::vector_mask
mxser_board::vector_mask is just a bitfield with bits set for all
available ports. We can obtain this value simply by
"BIT(brd->info->nports) - 1" directly in the ISR. So remove vector_mask
and simplify the code a bit.
Jiri Slaby [Fri, 18 Jun 2021 06:14:19 +0000 (08:14 +0200)]
mxser: extract port ISR
The ISR is terrible mix of letters. Let's extract the proper per-port
handling to a separate function called mxser_port_isr. This way, we can
actually see what both mxser_interrupt and mxser_port_isr do now.
Jiri Slaby [Fri, 18 Jun 2021 06:14:18 +0000 (08:14 +0200)]
mxser: cleanup LSR handling in mxser_receive_chars
mxser_interrupt is a bit confusing as it passes a pointer to LSR
variable ("status") to mxser_receive_chars. That function can indeed
update the LCR variable. Make this clear by not passing a variable by
reference, instead, return updated status. And change its type
accordingly -- to u8.
Note that the next patches will rework ISR so that even the definition
of status will be u8 soon too.
Jiri Slaby [Fri, 18 Jun 2021 06:14:17 +0000 (08:14 +0200)]
mxser: remove nonsense from ISR
request_irq was passed a board. There is no need to check dev_id and/or
find the appropriate board in the ISR. Neither check if board is NULL.
Remove this nonsense.
Jiri Slaby [Fri, 18 Jun 2021 06:14:13 +0000 (08:14 +0200)]
mxser: rename mxser_board::chip_flag to must_hwid
It is named IsMoxaMustChipFlag in the original Moxa's driver. Over the
time, it was reduced to chip_flag which doesn't make much sense. So dub
it again by what it is: must_hwid (mxser_must_get_hwid assigns to it).
Jiri Slaby [Fri, 18 Jun 2021 06:14:11 +0000 (08:14 +0200)]
mxser: cleanup Gpci_uart_info struct
The Gpci_uart_info structure contains up to tripled information and
chose weird types for the members. Deduplicate the members and choose
proper types for the data -- everything can be u8 except baud rate.
There, we use speed_t, which is used in this context.
Jiri Slaby [Fri, 18 Jun 2021 06:14:10 +0000 (08:14 +0200)]
mxser: integrate mxser.h into .c
As mxser.h serves only mxser.c, integrate the definitions into .c and
drop .h. There are some "semi-public" ioctl definitions, but they are
mostly go away in the next patches. If something, they would need to
live in an uapi header anyway.
Jiri Slaby [Fri, 18 Jun 2021 06:14:08 +0000 (08:14 +0200)]
mxser: renumber mxser_cards
After the ISA cards removal in the previous patch, the mxser_cards array
has holes in it. So renumber the array while updating PCI device's
driver_data.
Jiri Slaby [Fri, 18 Jun 2021 06:14:07 +0000 (08:14 +0200)]
mxser: drop ISA support
While mxser PCI cards are still around and produced (Moxa provided me
with two recently), ISA cards are obsolete for a long time. I haven't
seen anyone using the cards and the ISA code paths are barely tested.
Hence, remove ISA support from mxser driver so that we can clean the
driver up.
Ondrej Zary [Fri, 11 Jun 2021 20:19:39 +0000 (22:19 +0200)]
serial_cs: remove wrong GLOBETROTTER.cis entry
The GLOBETROTTER.cis entry in serial_cs matches more devices than
intended and breaks them. Remove it.
Example: # pccardctl info
PRODID_1="Option International
"
PRODID_2="GSM-Ready 56K/ISDN
"
PRODID_3="021
"
PRODID_4="A
"
MANFID=0013,0000
FUNCID=0
result:
pcmcia 0.0: Direct firmware load for cis/GLOBETROTTER.cis failed with error -2
The GLOBETROTTER.cis is nowhere to be found. There's GLOBETROTTER.cis.ihex at
https://netdev.vger.kernel.narkive.com/h4inqdxM/patch-axnet-cs-fix-phy-id-detection-for-bogus-asix-chip#post41
It's from completely diffetent card:
vers_1 4.1, "Option International", "GSM/GPRS GlobeTrotter", "001", "A"
MIPS: Malta: Enable magic multipliers for Super I/O UARTs
The SMSC FDC37M817 Super I/O chip has a configuration feature that lets
it support special UART divisor values of 32770 and 32769 for bit rates
of 230400 and 460800 bits per second respectively. Our 8250 driver core
provides support for these special divisors via the UPF_MAGIC_MULTIPLIER
flag, and YAMON firmware unconditionally configures the Super I/O chip
with these divisors enabled as well, so all we need to do in platform
setup for these bit rates to work is to set the flag.
serial: 8250: Handle custom baud rates in UPF_MAGIC_MULTIPLIER range
Handle custom baud rates requested in the UPF_MAGIC_MULTIPLIER range
with BOTHER. Currently matching is exact, that is if a baud rate that
is not either of clk/4 or clk/8 is requested, then we fall through to
the default case, which will just divide the clock rate by 16 times the
rate requested, round it to closest integer, and possibly yield even
worse results then if clamping to the extra baud rates was chosen.
So for example if we have the usual base rate of 115200 and request a
rate of 230399, then the fall-through divisor calculation will yield 1,
and consequently the baud rate of 115200 will be programmed even though
obviously the magic rate of 230400 would be more appropriate.
Make the selection of the magic rates range-qualified then and use clk/4
for rates from clk/6 up (assuming `serial8250_get_baud_rate' has already
rejected any rates too far beyond clk/4), and otherwise use clk/8 for
rates from clk/12 up, and finally fall through to the standard divisor
calculation. Explicitly void using the undocumented rate of 153600bps
and stick to documented divisor values only.
Support for magic baud rate divisors of 32770 and 32769 used with SMSC
Super I/O chips for extra baud rates of 230400 and 460800 respectively
where base rate is 115200[1] has been added around Linux 2.5.64, which
predates our repo history, but the origin could be identified as commit 2a717aad772f ("Merge with Linux 2.5.64.") with the old MIPS/Linux repo
also at: <git://git.kernel.org/pub/scm/linux/kernel/git/ralf/linux.git>.
Code that is now in `serial8250_do_get_divisor' was added back then to
`serial8250_get_divisor', but that code would only ever trigger if one
of the higher baud rates was actually requested, and that cannot ever
happen, because the earlier call to `serial8250_get_baud_rate' never
returns them. This is because it calls `uart_get_baud_rate' with the
maximum requested being the base rate, that is clk/16 or 115200 for SMSC
chips at their nominal clock rate.
Fix it then and allow UPF_MAGIC_MULTIPLIER baud rates to be selected, by
requesting the maximum baud rate of clk/4 rather than clk/16 if the flag
has been set. Also correct the minimum baud rate, observing that these
ports only support actual (non-magic) divisors of up to 32767 only.
References:
[1] "FDC37M81x, PC98/99 Compliant Enhanced Super I/O Controller with
Keyboard/Mouse Wake-Up", Standard Microsystems Corporation, Rev.
03/27/2000, Table 31 - "Baud Rates", p. 77
serial: 8250: Document SMSC Super I/O UART peculiarities
Contrary to what SMSC documentation says and unlike NS16C550A UARTs the
SMSC Super I/O IC claims compatibility with the SMSC UART implementation
does not support dividing the internal PLL clock by any divisor from 1
to 65535[1], with the exception of two magic divisors of 32769 and 32770
used respectively to select the high-speed data rates of 460800bps and
230400bps[2] if enabled on a port-by-port basis in with the Serial Port
Mode Register in the Device Configuration Space[3][4].
Instead empirical evidence indicates that the divisor, as stored in the
DLL and DLM register pair, has the range of 1 to 32767 only, and bit 7
of the DLM register (bit 15 of the divisor) effectively serves as a
selection bit for the prescaler from the base frequency of 7.3728MHz,
either 4 if the bit is 0, or 1 if the bit is 1 and high-speed operation
has been enabled with the Serial Port Mode Register.
So if high-speed operation has not been enabled, then say the values of
1 and 32769 (0x8001) written into the combined DLL and DLM register pair
both select the divisor of 1 and the baud rate of 115200bps.
[1] "FDC37M81x, PC98/99 Compliant Enhanced Super I/O Controller with
Keyboard/Mouse Wake-Up", Standard Microsystems Corporation, Rev.
03/27/2000, Section "Programmable Baud Rate Generator (and Divisor
Latches DLH, DLL)", p. 75
[2] same, Table 31 - "Baud Rates", p. 77
[3] same, Table 60 - "Serial Port 1, Logical Device 4 [Logical Device
Number = 0x04]", p. 153
[4] same, Table 61 - "Serial Port 2, Logical Device 5 [Logical Device
Number = 0x05]", p. 153
serial: 8250: Correct the clock for OxSemi PCIe devices
Oxford Semiconductor PCIe (Tornado) serial port devices are driven by a
fixed 62.5MHz clock input derived from the 100MHz PCI Express clock.
In the enhanced (650) mode, which we select in `autoconfig_has_efr' by
setting the ECB bit in the EFR register, and in the absence of clock
reconfiguration, which we currently don't do, the clock rate is divided
only by the oversampling rate of 16 as it is supplied to the baud rate
generator, yielding the baud base of 3906250. This comes from the reset
values of the TCR and MCR[7] registers which are both zero[1][2][3][4],
choosing the oversampling rate of 16 and the normal (divide by 1) baud
rate generator prescaler respectively. This is the rate that is divided
by the value held in the divisor latch to determine the baud rate used.
Replace the incorrect baud base of 4000000 with the right value of 3906250 then.
References:
[1] "OXPCIe200 PCI Express Multi-Port Bridge", Oxford Semiconductor,
Inc., DS-0045, 10 Nov 2008, Section "Reset Configuration", p. 72
[2] "OXPCIe952 PCI Express Bridge to Dual Serial & Parallel Port",
Oxford Semiconductor, Inc., DS-0046, Mar 06 08, Section "Reset
Configuration", p. 27
[3] "OXPCIe954 PCI Express Bridge to Quad Serial Port", Oxford
Semiconductor, Inc., DS-0047, Feb 08, Section "Reset Configuration",
p. 28
[4] "OXPCIe958 PCI Express Bridge to Octal Serial Port", Oxford
Semiconductor, Inc., DS-0048, Feb 08, Section "Reset Configuration",
p. 28
serial: 8250: Dissociate 4MHz Titan ports from Oxford ports
Oxford Semiconductor PCIe (Tornado) serial port devices have their baud
base set incorrectly, however their `pciserial_board' entries have been
reused for Titan serial port devices. Define own entries for the latter
devices then, carrying over the settings, so that Oxford entries can be
fixed.
Jiri Slaby [Thu, 10 Jun 2021 09:02:47 +0000 (11:02 +0200)]
tty: make use of tty_get_{char,frame}_size
In the previous patch, we introduced tty_get_char_size() and
tty_get_frame_size() for computing character and frame sizes,
respectively. Here, we make use of them in various tty drivers where
applicable.
The stats look nice: 12 insertions, 169 deletions.
Cc: Arnd Bergmann <arnd@arndb.de> Cc: David Lin <dtwlin@gmail.com> Cc: Johan Hovold <johan@kernel.org> Cc: Alex Elder <elder@kernel.org> Cc: Shawn Guo <shawnguo@kernel.org> Cc: Sascha Hauer <s.hauer@pengutronix.de> Cc: Andy Gross <agross@kernel.org> Cc: Bjorn Andersson <bjorn.andersson@linaro.org> Cc: Maxime Coquelin <mcoquelin.stm32@gmail.com> Cc: Alexandre Torgue <alexandre.torgue@foss.st.com> Cc: Oliver Neukum <oneukum@suse.com> Acked-by: Alex Elder <elder@kernel.org> Signed-off-by: Jiri Slaby <jslaby@suse.cz> Link: https://lore.kernel.org/r/20210610090247.2593-4-jslaby@suse.cz Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Jiri Slaby [Thu, 10 Jun 2021 09:02:46 +0000 (11:02 +0200)]
cypress_m8: switch data_bits to real character bits
Make data_bits what it really is. Assign proper bit counts to data_bits
instead of magic 0..3. There are two reasons:
1) it's clear what we store there, and
2) it will make the transition to tty_tty_get_char_size() in the next
patch easier.
Jiri Slaby [Thu, 10 Jun 2021 09:02:45 +0000 (11:02 +0200)]
mxs-auart: redefine AUART_LINECTRL_WLEN to accept bits count
It's a bit illogical to take magic constants in AUART_LINECTRL_WLEN.
Like in auart_console_get_options(), 2 means 7 bits. Switch
AUART_LINECTRL_WLEN to accept bit length, i.e. let it subtract 5 from
the parameter before doing the logic.
This will ease conversion from CSIZE to bits in mxs_auart_settermios()
in the next patch.
Jiri Slaby [Thu, 10 Jun 2021 09:02:44 +0000 (11:02 +0200)]
tty: make tty_get_{char,frame}_size available
Many tty drivers contain code to compute bits count depending on termios
cflags. So extract this code from serial core to two separate tty helper
functions:
* tty_get_char_size -- only size of a character, without flags,
* tty_get_frame_size -- complete size of a frame including flags.
In the next patch, calls to these new functions replace many copies of
this code.
Note that we accept only cflag as a parameter. That's because some
callers like pch_uart_startup or sunsab_console_setup don't have at hand
termios which we could pass around.
Cc: Joe Perches <joe@perches.com> Cc: Johan Hovold <johan@kernel.org> Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com> Signed-off-by: Jiri Slaby <jslaby@suse.cz> Link: https://lore.kernel.org/r/20210610090247.2593-1-jslaby@suse.cz Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Jiri Slaby [Thu, 10 Jun 2021 09:03:07 +0000 (11:03 +0200)]
ipwireless: remove unused ipw_tty::closing
It's only set, but never read because commit 01261cb94318 ("tty:
ipwireless: Remove tty->closing abort from ipw_open()") removed last use
of tty::closing.
Erwan Le Ray [Thu, 10 Jun 2021 10:00:20 +0000 (12:00 +0200)]
serial: stm32: defer probe for dma devices
Defer device probe operation when a DMA channel request probe deferral.
With this change both DMA channels are acquired before DMA channels are
configured. Once no probe deferral is expected, DMAs are
configured and any failure in their configuration will make the driver
to fallback to interrupt mode as prior this change.
Mikko Perttunen [Wed, 9 Jun 2021 15:56:55 +0000 (18:56 +0300)]
serial: tegra-tcu: Reorder channel initialization
Request the RX mailbox only after initializing the UART data
structures. Otherwise it can rarely happen that the receive
callback is called before the UART is ready.
serial: sh-sci: Stop dmaengine transfer in sci_stop_tx()
Stop dmaengine transfer in sci_stop_tx(). Otherwise, the following
message is possible output when system enters suspend and while
transferring data, because clearing TIE bit in SCSCR is not able to
stop any dmaengine transfer.
sh-sci e6550000.serial: ttySC1: Unable to drain transmitter
Note that this driver has already used some #ifdef in the .c file
so that this patch also uses #ifdef to fix the issue. Otherwise,
build errors happens if the CONFIG_SERIAL_SH_SCI_DMA is disabled.
Linus Torvalds [Sun, 13 Jun 2021 19:41:47 +0000 (12:41 -0700)]
Merge tag 'perf-tools-fixes-for-v5.13-2021-06-13' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux
Pull perf tools fixes from Arnaldo Carvalho de Melo:
- Correct buffer copying when peeking events
- Sync cpufeatures/disabled-features.h header with the kernel sources
* tag 'perf-tools-fixes-for-v5.13-2021-06-13' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux:
tools headers cpufeatures: Sync with the kernel sources
perf session: Correct buffer copying when peeking events
Linus Torvalds [Sun, 13 Jun 2021 19:32:59 +0000 (12:32 -0700)]
Merge tag 'nfs-for-5.13-3' of git://git.linux-nfs.org/projects/trondmy/linux-nfs
Pull NFS client bugfixes from Trond Myklebust:
"Highlights include:
Stable fixes:
- Fix use-after-free in nfs4_init_client()
Bugfixes:
- Fix deadlock between nfs4_evict_inode() and nfs4_opendata_get_inode()
- Fix second deadlock in nfs4_evict_inode()
- nfs4_proc_set_acl should not change the value of NFS_CAP_UIDGID_NOMAP
- Fix setting of the NFS_CAP_SECURITY_LABEL capability"
* tag 'nfs-for-5.13-3' of git://git.linux-nfs.org/projects/trondmy/linux-nfs:
NFSv4: Fix second deadlock in nfs4_evict_inode()
NFSv4: Fix deadlock between nfs4_evict_inode() and nfs4_opendata_get_inode()
NFS: FMODE_READ and friends are C macros, not enum types
NFS: Fix a potential NULL dereference in nfs_get_client()
NFS: Fix use-after-free in nfs4_init_client()
NFS: Ensure the NFS_CAP_SECURITY_LABEL capability is set when appropriate
NFSv4: nfs4_proc_set_acl needs to restore NFS_CAP_UIDGID_NOMAP on error.
Linus Torvalds [Sun, 13 Jun 2021 19:25:33 +0000 (12:25 -0700)]
Merge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Pull SCSI fixes from James Bottomley:
"Four reasonably small fixes to the core for scsi host allocation
failure paths.
The root problem is that we're not freeing the memory allocated by
dev_set_name(), which involves a rejig of may of the free on error
paths to do put_device() instead of kfree which, in turn, has several
other knock on ramifications and inspection turned up a few other
lurking bugs"
* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
scsi: core: Only put parent device if host state differs from SHOST_CREATED
scsi: core: Put .shost_dev in failure path if host state changes to RUNNING
scsi: core: Fix failure handling of scsi_add_host_with_dma()
scsi: core: Fix error handling of scsi_host_alloc()
Linus Torvalds [Sat, 12 Jun 2021 20:57:49 +0000 (13:57 -0700)]
Merge tag 'riscv-for-linus-5.13-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux
Pull RISC-V fixes from Palmer Dabbelt:
- A pair of XIP fixes: one to fix alternatives, and one to turn off the
rest of the features that require code modification
- A fix to a type that was causing some alternatives to break
- A build fix for BUILTIN_DTB
* tag 'riscv-for-linus-5.13-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:
riscv: Fix BUILTIN_DTB for sifive and microchip soc
riscv: alternative: fix typo in macro name
riscv: code patching only works on !XIP_KERNEL
riscv: xip: support runtime trap patching
Feng Tang [Fri, 11 Jun 2021 01:54:42 +0000 (09:54 +0800)]
mm: relocate 'write_protect_seq' in struct mm_struct
0day robot reported a 9.2% regression for will-it-scale mmap1 test
case[1], caused by commit 57efa1fe5957 ("mm/gup: prevent gup_fast from
racing with COW during fork").
Further debug shows the regression is due to that commit changes the
offset of hot fields 'mmap_lock' inside structure 'mm_struct', thus some
cache alignment changes.
From the perf data, the contention for 'mmap_lock' is very severe and
takes around 95% cpu cycles, and it is a rw_semaphore
Before commit 57efa1fe5957 adds the 'write_protect_seq', it happens to
have a very optimal cache alignment layout, as Linus explained:
"and before the addition of the 'write_protect_seq' field, the
mmap_sem was at offset 120 in 'struct mm_struct'.
Which meant that count and owner were in two different cachelines,
and then when you have contention and spend time in
rwsem_down_write_slowpath(), this is probably *exactly* the kind
of layout you want.
Because first the rwsem_write_trylock() will do a cmpxchg on the
first cacheline (for the optimistic fast-path), and then in the
case of contention, rwsem_down_write_slowpath() will just access
the second cacheline.
Which is probably just optimal for a load that spends a lot of
time contended - new waiters touch that first cacheline, and then
they queue themselves up on the second cacheline."
After the commit, the rw_semaphore is at offset 128, which means the
'count' and 'owner' fields are now in the same cacheline, and causes
more cache bouncing.
Currently there are 3 "#ifdef CONFIG_XXX" before 'mmap_lock' which will
affect its offset:
The layout above is on 64 bits system with 0day's default kernel config
(similar to RHEL-8.3's config), in which all these 3 options are 'y'.
And the layout can vary with different kernel configs.
Relayouting a structure is usually a double-edged sword, as sometimes it
can helps one case, but hurt other cases. For this case, one solution
is, as the newly added 'write_protect_seq' is a 4 bytes long seqcount_t
(when CONFIG_DEBUG_LOCK_ALLOC=n), placing it into an existing 4 bytes
hole in 'mm_struct' will not change other fields' alignment, while
restoring the regression.
Link: https://lore.kernel.org/lkml/20210525031636.GB7744@xsang-OptiPlex-9020/ Reported-by: kernel test robot <oliver.sang@intel.com> Signed-off-by: Feng Tang <feng.tang@intel.com> Reviewed-by: John Hubbard <jhubbard@nvidia.com> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> Cc: Peter Xu <peterx@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Linus Torvalds [Sat, 12 Jun 2021 19:34:49 +0000 (12:34 -0700)]
Merge tag 'usb-5.13-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
Pull USB fixes from Greg KH:
"Here are a number of tiny USB fixes for 5.13-rc6.
There are more than I would normally like, but there's been a bunch of
people banging on the gadget and dwc3 and typec code recently for I
think an Android release, which has resulted in a number of small
fixes. It's nice to see companies send fixes upstream for this type of
work, a notable change from years ago.
Anyway, fixes in here are:
- usb-serial device id updates
- usb-serial cp210x driver fixes for broken firmware versions
- typec fixes for crazy charging devices and other reported problems
- dwc3 fixes for reported problems found
- gadget fixes for reported problems
- tiny xhci fixes
- other small fixes for reported issues.
- revert of a problem fix found by linux-next testing
All of these have passed 0-day and linux-next testing with no reported
problems (the revert for the found linux-next build problem included)"
* tag 'usb-5.13-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (44 commits)
Revert "usb: gadget: fsl: Re-enable driver for ARM SoCs"
usb: typec: mux: Fix copy-paste mistake in typec_mux_match
usb: typec: ucsi: Clear PPM capability data in ucsi_init() error path
usb: gadget: fsl: Re-enable driver for ARM SoCs
usb: typec: wcove: Use LE to CPU conversion when accessing msg->header
USB: serial: cp210x: fix CP2102N-A01 modem control
USB: serial: cp210x: fix alternate function for CP2102N QFN20
usb: misc: brcmstb-usb-pinmap: check return value after calling platform_get_resource()
usb: dwc3: ep0: fix NULL pointer exception
usb: gadget: eem: fix wrong eem header operation
usb: typec: intel_pmc_mux: Put ACPI device using acpi_dev_put()
usb: typec: intel_pmc_mux: Add missed error check for devm_ioremap_resource()
usb: typec: intel_pmc_mux: Put fwnode in error case during ->probe()
usb: typec: tcpm: Do not finish VDM AMS for retrying Responses
usb: fix various gadget panics on 10gbps cabling
usb: fix various gadgets null ptr deref on 10gbps cabling.
usb: pci-quirks: disable D3cold on xhci suspend for s2idle on AMD Renoir
usb: f_ncm: only first packet of aggregate needs to start timer
USB: f_ncm: ncm_bitrate (speed) is unsigned
MAINTAINERS: usb: add entry for isp1760
...
Linus Torvalds [Sat, 12 Jun 2021 19:13:55 +0000 (12:13 -0700)]
Merge tag 'char-misc-5.13-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc
Pull char/misc driver fixes from Greg KH:
"Here are some small misc driver fixes for 5.13-rc6 that fix some
reported problems:
- Tiny phy driver fixes for reported issues
- rtsx regression for when the device suspended
- mhi driver fix for a use-after-free
All of these have been in linux-next for a few days with no reported
issues"
* tag 'char-misc-5.13-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:
misc: rtsx: separate aspm mode into MODE_REG and MODE_CFG
bus: mhi: pci-generic: Fix hibernation
bus: mhi: pci_generic: Fix possible use-after-free in mhi_pci_remove()
bus: mhi: pci_generic: T99W175: update channel name from AT to DUN
phy: Sparx5 Eth SerDes: check return value after calling platform_get_resource()
phy: ralink: phy-mt7621-pci: drop 'of_match_ptr' to fix -Wunused-const-variable
phy: ti: Fix an error code in wiz_probe()
phy: phy-mtk-tphy: Fix some resource leaks in mtk_phy_init()
phy: cadence: Sierra: Fix error return code in cdns_sierra_phy_probe()
phy: usb: Fix misuse of IS_ENABLED
Linus Torvalds [Sat, 12 Jun 2021 19:06:24 +0000 (12:06 -0700)]
Merge tag 'pinctrl-v5.13-2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl
Pull pin control fixes from Linus Walleij:
- Fix some documentation warnings for Allwinner
- Fix duplicated GPIO groups on Qualcomm SDX55
- Fix a double enablement bug in the Ralink driver
- Fix the Qualcomm SC8180x Kconfig so the driver can be selected.
* tag 'pinctrl-v5.13-2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl:
pinctrl: qcom: Make it possible to select SC8180x TLMM
pinctrl: ralink: rt2880: avoid to error in calls is pin is already enabled
pinctrl: qcom: Fix duplication in gpio_groups
pinctrl: aspeed: Fix minor documentation error
Linus Torvalds [Sat, 12 Jun 2021 18:59:58 +0000 (11:59 -0700)]
Merge tag 'block-5.13-2021-06-12' of git://git.kernel.dk/linux-block
Pull block fixes from Jens Axboe:
"A few fixes that should go into 5.13:
- Fix a regression deadlock introduced in this release between open
and remove of a bdev (Christoph)
- Fix an async_xor md regression in this release (Xiao)
- Fix bcache oversized read issue (Coly)"
* tag 'block-5.13-2021-06-12' of git://git.kernel.dk/linux-block:
block: loop: fix deadlock between open and remove
async_xor: check src_offs is not NULL before updating it
bcache: avoid oversized read request in cache missing code path
bcache: remove bcache device self-defined readahead
Linus Torvalds [Sat, 12 Jun 2021 18:53:20 +0000 (11:53 -0700)]
Merge tag 'io_uring-5.13-2021-06-12' of git://git.kernel.dk/linux-block
Pull io_uring fixes from Jens Axboe:
"Just an API change for the registration changes that went into this
release. Better to get it sorted out now than before it's too late"
* tag 'io_uring-5.13-2021-06-12' of git://git.kernel.dk/linux-block:
io_uring: add feature flag for rsrc tags
io_uring: change registration/upd/rsrc tagging ABI
Linus Torvalds [Sat, 12 Jun 2021 18:41:28 +0000 (11:41 -0700)]
Merge tag 'sched-urgent-2021-06-12' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler fixes from Ingo Molnar:
"Misc fixes:
- Fix performance regression caused by lack of intended batching of
RCU callbacks by over-eager NOHZ-full code.
- Fix cgroups related corruption of load_avg and load_sum metrics.
- Three fixes to fix blocked load, util_sum/runnable_sum and util_est
tracking bugs"
* tag 'sched-urgent-2021-06-12' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/fair: Fix util_est UTIL_AVG_UNCHANGED handling
sched/pelt: Ensure that *_sum is always synced with *_avg
tick/nohz: Only check for RCU deferred wakeup on user/guest entry when needed
sched/fair: Make sure to update tg contrib for blocked load
sched/fair: Keep load_avg and load_sum synced
Linus Torvalds [Sat, 12 Jun 2021 18:34:49 +0000 (11:34 -0700)]
Merge tag 'perf-urgent-2021-06-12' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf fixes from Ingo Molnar:
"Misc fixes:
- Fix the NMI watchdog on ancient Intel CPUs
- Remove a misguided, NMI-unsafe KASAN callback from the NMI-safe
irq_work path used by perf.
- Fix uncore events on Ice Lake servers.
- Someone booted maxcpus=1 on an SNB-EP, and the uncore driver
emitted warnings and was probably buggy. Fix it.
- KCSAN found a genuine data race in the core perf code. Somewhat
ironically the bug was introduced through a recent race fix. :-/
In our defense, the new race window was much more narrow. Fix it"
* tag 'perf-urgent-2021-06-12' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/nmi_watchdog: Fix old-style NMI watchdog regression on old Intel CPUs
irq_work: Make irq_work_queue() NMI-safe again
perf/x86/intel/uncore: Fix M2M event umask for Ice Lake server
perf/x86/intel/uncore: Fix a kernel WARNING triggered by maxcpus=1
perf: Fix data race between pin_count increment/decrement
Linus Torvalds [Sat, 12 Jun 2021 18:10:28 +0000 (11:10 -0700)]
Merge tag 'objtool-urgent-2021-06-12' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull objtool fixes from Ingo Molnar:
"Two objtool fixes:
- fix a bug that corrupts the code by mistakenly rewriting
conditional jumps
- fix another bug generating an incorrect ELF symbol table
during retpoline rewriting"
* tag 'objtool-urgent-2021-06-12' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
objtool: Only rewrite unconditional retpoline thunk calls
objtool: Fix .symtab_shndx handling for elf_create_undef_symbol()
Alexandre Ghiti [Fri, 4 Jun 2021 12:06:39 +0000 (14:06 +0200)]
riscv: Fix BUILTIN_DTB for sifive and microchip soc
Fix BUILTIN_DTB config which resulted in a dtb that was actually not
built into the Linux image: in the same manner as Canaan soc does,
create an object file from the dtb file that will get linked into the
Linux image.
Linus Torvalds [Sat, 12 Jun 2021 00:05:03 +0000 (17:05 -0700)]
Merge tag 'trace-v5.13-rc5-2' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace
Pull tracing fixes from Steven Rostedt:
- Fix the length check in the temp buffer filter
- Fix build failure in bootconfig tools for "fallthrough" macro
- Fix error return of bootconfig apply_xbc() routine
* tag 'trace-v5.13-rc5-2' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
tracing: Correct the length check which causes memory corruption
ftrace: Do not blindly read the ip address in ftrace_bug()
tools/bootconfig: Fix a build error accroding to undefined fallthrough
tools/bootconfig: Fix error return code in apply_xbc()
projects / thirdparty / kernel / linux.git / log
